Fish oil · Heart & health
EPA and DHA, the omega-3 fats of oily fish. They lower triglycerides, which kept falling with dose in the trials, and blood pressure by a little, most at 2 to 3 g a day.
Thiago’s take: Core · 325 mg EPA + DHA every morningEPA and DHA are the long-chain omega-3 fats in oily fish, and in fish, krill and algal oils. A supplement is measured by its EPA and DHA, not by the weight of its oil: two softgels holding 2,500 mg of salmon oil carry 325 mg of the two.
S-tier. Most people don’t eat enough oily fish. This covers it.
Morning with breakfast
Training 30 minutes before
Evening with dinner
Bed 30 minutes before
3 Strong4 Preliminary
Strong: at least one pooled review or three randomised trials among the counted studies, and at least three in four of them better. Preliminary: everything else. Better is green in a Strong finding and ochre in a Preliminary one.
Better means a statistically significant change in the claimed direction at a normal dose. Each study counts once: a pooled review speaks for the trials inside it.
72,598 people, Wang 2023 · the more taken, the bigger the drop
5 of 5 studies better
Every review found triglycerides lower with omega-3, and the more taken, the bigger the drop: about 19 mg/dL at 1 g a day of EPA and DHA, and 43 at 2 g. The drop is largest in people whose triglycerides start high.
Strong
Heart
What these 6 studies found. The ring counts 5 of them: the other is a trial pooled in Wang 2023, counted through it. Pooled reviews: 5 better. Pooled in Wang 2023: 1 better.
Triglycerides fell with dose: about 19 mg/dL lower than control at 1 g a day of EPA and DHA, 43 at 2 g and 69 at 3 g. Its trials include prescription EPA, JELIS among them, and the curve held without the trials of EPA or DHA alone.
Wang 2023 Journal of the American Heart Association
In adults with high triglycerides, about 40 mg/dL lower than placebo (19 of its 32 trials), and about 30 lower when added to a statin (14 trials); its largest trial, STRENGTH, gave a prescription omega-3 at 4 g a day against corn oil to 13,078 people.
Yang 2022 Frontiers in Nutrition
In 12,536 people at high cardiovascular risk with diabetes or raised blood sugar, in a trial that also tested insulin, 1 g a day of omega-3 ethyl esters lowered triglycerides by 14.5 mg/dL more than placebo.
Bosch 2012 New England Journal of Medicine
In mixed groups of adults, many with heart disease, polycystic ovary syndrome or kidney disease, triglycerides about 17 mg/dL lower than in the control groups, and HDL cholesterol slightly higher.
Chen 2026 Nutrition, Metabolism and Cardiovascular Diseases
In adults aged 30 to 70 doing exercise training, adding omega-3 lowered triglycerides by 0.10 mmol/L more than training alone, and raised LDL cholesterol by 0.14 mmol/L.
Khalafi 2025 Clinical Nutrition ESPEN
In overweight and obese adults, triglycerides lower than placebo (9 of its 11 trials), with no difference in weight; it calls its trials few and of poor quality.
Zhang 2017 The Journal of Nutrition, Health & Aging
Found with a PubMed search: omega-3, n-3, fish oil, EPA or DHA in the title, triglycerides in the title or abstract, reviews, and trials with triglycerides or lipids in the title. 418 results on 4 Oct 2026.
4,973 people, Zhang 2022 · the biggest drop at 2 to 3 g a day
3 of 4 studies better
Omega-3 lowers blood pressure a little: about 1.5 to 2.6 mmHg systolic across adults, most at 2 to 3 g a day, and more in people with untreated high blood pressure. Added to exercise training, it lowered blood pressure by about 4 mmHg more than training alone; the largest trial, of 2,157 people aged 70 and over taking 1 g a day, found no significant drop.
Strong
Heart
What these 4 studies found. Pooled reviews: 3 better. Trials: 1 no change.
Blood pressure fell most at 2 to 3 g a day of EPA and DHA, about 2.6 mmHg systolic and 1.6 to 1.8 diastolic below control, with bigger drops in people with high blood pressure or high blood fats and in trials of adults aged 45 and over.
Zhang 2022 Journal of the American Heart Association
Systolic pressure 1.5 mmHg and diastolic 1.0 lower than placebo: 4.5 and 3.1 lower in people with untreated high blood pressure, 1.3 and 0.6 with normal blood pressure. The drop came from supplements; its 11 results from enriched foods and fish showed none.
Miller 2014 American Journal of Hypertension
In 2,157 adults aged 70 and over in good health, in a trial that also tested vitamin D and strength training, 1 g a day of omega-3 for three years left systolic pressure 0.8 mmHg and diastolic 0.5 lower than without omega-3, neither significant.
Bischoff-Ferrari 2020 JAMA
In adults aged 30 to 70 doing exercise training, adding omega-3 lowered systolic and diastolic pressure each by about 4 mmHg more than training alone.
Khalafi 2025 Clinical Nutrition ESPEN
Found with a PubMed search: omega-3, n-3, fish oil, EPA or DHA in the title, blood pressure or hypertension in the title or abstract, trials and reviews. 296 results on 4 Oct 2026.
4,601 people, Li 2014 · ill and healthy adults, 17 of its 68 trials healthy
1 of 3 studies better
In people with a long-term illness, omega-3 lowered CRP and IL-6 by about a fifth; in healthy people it lowered CRP by about a sixth, but not clearly IL-6. A year-long trial of 1,561 adults aged 50 and over taking 1 g a day lowered neither, and added to exercise training, omega-3 did not lower IL-6 more than training alone.
Preliminary
Inflammation
What these 3 studies found. Pooled reviews: 1 better, 1 no change. Trials: 1 no change.
In people with a long-term illness other than autoimmune disease (48 of its 68 trials), CRP and IL-6 about 18 and 20% lower; in healthy people (17 trials), CRP about 16% lower and IL-6 not significantly. The CRP drop was significant against placebo oils rich in linoleic acid, such as corn or sunflower oil, not against oleic-acid oils such as olive oil.
Li 2014 PLOS ONE
In 1,561 adults aged 50 and over, in a trial that also tested vitamin D, 1 g a day for a year lowered neither IL-6 nor CRP compared with placebo.
Costenbader 2019 Clinical Chemistry
In adults aged 30 to 70 doing exercise training, adding omega-3 lowered TNF-α but left IL-6 no lower than training alone; CRP was not among its outcomes.
Khalafi 2025 Clinical Nutrition ESPEN
Found with a PubMed search: omega-3, n-3, fish oil, EPA or DHA in the title, C-reactive protein, interleukin-6 or inflammatory markers in the title or abstract, reviews, and trials with inflammation, CRP or cytokines in the title. 318 results on 4 Oct 2026.
301 people, Lv 2020 · less soreness, but below a clinically important size
3 of 3 studies better
Two pooled reviews, mostly of young men new to eccentric exercise, found less soreness in the days after with omega-3, though the larger found the drop smaller than a clinically important amount. A trial in 32 active adults found less soreness at 2,800 mg a day of EPA and DHA.
Strong
Recovery
What these 3 studies found. Pooled reviews: 2 better. Trials: 1 better.
In adults new to eccentric exercise, soreness two days after was 0.93 points lower than placebo on a 10-point scale, under the 1.4 points it counts as clinically important; strength and range of motion did not differ. Its 12 trials include a purslane extract and a DHA blend with flavonoids.
Lv 2020 BioMed Research International
In young men, mostly untrained, peak soreness after eccentric exercise was lower than placebo, a moderate-to-large effect, with lower creatine kinase, less swelling and better strength and range of motion; it calls the wider evidence, 43 studies, equivocal.
Yaghoobi 2026 Nutrients
In its 2,800 mg and placebo arms (16 of 32 active men and women), 2,800 mg EPA + DHA a day for about 7.5 weeks left less soreness than safflower oil 24 and 72 hours after heavy squats; the 4,200 mg arm, above the usual range, had the least.
VanDusseldorp 2020 Nutrients
Found with a PubMed search: omega-3, n-3, fish oil, EPA or DHA in the title, muscle damage, soreness, recovery or eccentric exercise in the title or abstract, trials and reviews. 91 results on 4 Oct 2026.
162,796 people, Abdelhamid 2020 · slightly fewer coronary events, low certainty
1 of 2 studies better
The largest review found slightly fewer coronary events with omega-3, about 9% fewer at low certainty, and 8% fewer cardiovascular deaths, which it grades little or no effect. A review of high-dose, mostly EPA omega-3 found no clear drop in heart attacks or cardiovascular deaths, and a trial of 12,536 people at high risk taking 1 g a day saw no fewer cardiovascular deaths.
Preliminary
Heart
What these 3 studies found. The ring counts 2 of them: the other is a trial pooled in Abdelhamid 2020, counted through it. Pooled reviews: 1 better, 1 no change. Pooled in Abdelhamid 2020: 1 no change.
In adults at varying cardiovascular risk, more EPA and DHA than placebo or usual diet left cardiovascular deaths 8% lower (29 of its 86 trials), which it grades little or no effect, and coronary events 9% lower (32 trials), a slight drop of low certainty. Most of its trials gave capsules, some enriched foods or dietary advice, and its 86 include trials of plant omega-3 (ALA).
Abdelhamid 2020 Cochrane Database of Systematic Reviews
In people with heart disease or at high risk, high-dose EPA-dominant omega-3 (at least 1.8 g a day, at least half EPA, 4 g a day of icosapent ethyl, a purified EPA, among it), against placebo or in open-label trials, lowered neither heart attacks nor cardiovascular deaths overall; heart attacks were lower once STRENGTH, its one mixed EPA + DHA outcome trial, was left out.
Faheem 2026 American Journal of Cardiovascular Drugs
In 12,536 people at high cardiovascular risk with diabetes or raised blood sugar, 1 g a day of omega-3 ethyl esters for a median 6.2 years left cardiovascular deaths no lower than placebo, in a trial that also tested insulin.
Bosch 2012 New England Journal of Medicine
Found with a PubMed search: omega-3, n-3, fish oil, EPA, DHA or icosapent ethyl in the title, heart attack, coronary disease or cardiovascular deaths or events in the title or abstract, reviews, and trials with the heart or blood vessels in the title. 373 results on 4 Oct 2026.
49,757 people, Brainard 2020 · little or no effect, moderate-quality evidence
1 of 2 studies better
The largest review found omega-3 raised scores on a 30-point thinking test by a tenth of a point, which it calls little or no effect, and did not prevent new cognitive impairment. A review of adults aged 40 and over without dementia found no change in memory or overall thinking, and the largest trial, in 2,157 people aged 70 and over, found no change in cognitive scores over three years.
Preliminary
Cognition
What these 3 studies found. The ring counts 2 of them: the other is a trial pooled in Suh 2024, counted through it. Pooled reviews: 1 better, 1 no change. Pooled in Suh 2024: 1 no change.
In adults, in trials of at least 24 weeks, long-chain omega-3 against lower intake raised Mini-Mental State scores by 0.10 of 30 points (13 of its 38 trials), which it calls little or no effect, and did not lower new cognitive impairment. Its trials also test plant omega-3, omega-6 and total polyunsaturated fat.
Brainard 2020 Journal of the American Medical Directors Association
In adults aged 40 and over without dementia, omega-3 changed neither overall thinking nor memory, attention, executive function, processing speed or visuospatial skill against placebo; a dose analysis saw executive function rise in the first 12 months above 500 mg a day. Its 24 trials include four of plant omega-3 (ALA) and three of DHA alone.
Suh 2024 BMC Medicine
In 2,157 adults aged 70 and over in good health, in a trial that also tested vitamin D and strength training, 1 g a day of omega-3 for three years left cognitive scores (MoCA) no different from placebo.
Bischoff-Ferrari 2020 JAMA
Found with a PubMed search: omega-3, n-3, fish oil, EPA or DHA in the title, cognition or memory in the title or abstract, reviews, and trials with cognition or memory in the title. 188 results on 4 Oct 2026.
2,070 people, Deng 2023 · less pain and better function in osteoarthritis
2 of 3 studies better
In osteoarthritis, a review of nine trials found a small drop in pain with omega-3, though its largest trial, of 1,398 older adults with knee pain, found none over five years. In rheumatoid arthritis, the newest review found fewer tender joints, and an earlier one, at 2.7 g a day or more, did not.
Preliminary
Joints
What these 4 studies found. The ring counts 3 of them: the other is a trial pooled in Deng 2023, counted through it. Pooled reviews: 2 better, 1 no change. Pooled in Deng 2023: 1 no change.
In 2,070 people with osteoarthritis, mostly of the knee or hip, pain was lower than with placebo (SMD −0.29) and function better. Its nine trials include two green-lipped mussel extracts, krill oil and a branded food supplement.
Deng 2023 Journal of Orthopaedic Surgery and Research
In 1,398 older US adults with chronic knee pain, in a trial that also tested vitamin D, 1 g a day of omega-3 for about 5 years left knee pain, stiffness and function no different from placebo.
MacFarlane 2020 Arthritis & Rheumatology
In 1,018 people with rheumatoid arthritis, fewer tender joints with omega-3 than in the control groups; disease activity scores (DAS28), CRP and ESR fell slightly but not significantly.
Wang 2024 Clinical Rheumatology
In people with rheumatoid arthritis taking at least 2.7 g a day for 3 months or more, tender and swollen joints were not significantly fewer than with placebo, though NSAID use fell.
Lee 2012 Archives of Medical Research
Found with a PubMed search: omega-3, n-3, fish oil, EPA or DHA in the title, arthritis, osteoarthritis, rheumatoid or joint pain in the title or abstract, trials and reviews. 82 results on 4 Oct 2026.
Guidelines set about 250 to 500 mg EPA + DHA a day, and most people eat under 200 mg. The findings’ effects came at higher doses: over 1,500 mg a day for triglycerides, blood pressure and inflammation, by one review, and blood pressure fell most at 2,000 to 3,000 mg. Thiago’s 325 mg sits in the 250 to 500 mg most guidelines name, and below the trials’ band.
Daily1,000 to 4,000 mg EPA + DHA, the trials’ band
Sources: Harris 2009 · Calder 2025 · Lewis 2025 · Zhang 2022 · Wang 2023
What people notice, what rises with the dose, and what the common worries turn out to be.
Across 90 trials, a fishy taste had about three and a half times the odds it had on placebo or standard treatment, and diarrhoea about a quarter higher odds. In trials of prescription EPA + DHA capsules, burping and nausea were more common too.
Sources: Chang 2023 · Chang 2018
An irregular heartbeat. In seven large heart trials of 81,210 people, average age 65, it was 12% more likely at up to 1,000 mg a day and 49% more likely above that, where the trials include prescription EPA (REDUCE-IT); a 2026 review of 35 trials in people aged 50 and over found no rise at lower doses, and a rise only above 1,500 mg a day in people at high cardiovascular risk (odds ratio 1.43).
Sources: Gencer 2021 · Abuknesha 2026
In 11 trials of 120,643 people, bleeding, including in the brain and gut, was no more common on omega-3. A review of 90 trials found higher odds of a bleeding tendency (odds ratio 1.26), a signal its subgroups tied to prescription products.
Sources: Javaid 2024 · Chang 2023
Warfarin or another blood thinner. In the records of 573 warfarin users, the 145 taking fish or krill oil showed no change in clotting control or bleeding, but the doctor managing a blood thinner should know.
Source: Pryce 2016
Fish oil’s natural form, and Thiago’s. In a 2010 study of 72 people, a concentrate turned back into triglycerides delivered 124% as much EPA + DHA as natural fish oil.
This pageA fish oil concentrate. The same study found it delivered 73% as much as natural fish oil, a 2015 trial found blood levels no different after four weeks, and it is absorbed far better with a fatty meal.
Krill oil carries its EPA and DHA largely as phospholipids. At similar doses, two trials found it raised blood levels no more than fish oil.
Made from algae, with no fish, so vegan. In a 2008 trial, 600 mg of DHA a day from algal-oil capsules raised blood DHA as much as the same from cooked salmon.
Icosapent ethyl, a purified EPA sold as a prescription drug, given at 4 g a day for raised triglycerides.
EPA and DHA work from inside cell membranes, where they change the signals cells make.
EPA and DHA come from oily fish, and from fish, krill and algal oils. The body makes little of them from ALA, a shorter omega-3.
Taken daily, they reach the blood within days, immune cells over months and body fat over more than a year.
In cell membranes they change the mix of fats, so cells make fewer prostaglandins and leukotrienes from arachidonic acid, an omega-6 fat. EPA’s own versions are weaker.
EPA and DHA are also made into resolvins, protectins and maresins, which help bring inflammation to an end.
Sources: Lewis 2025 · Yurko-Mauro 2015 · Harris 2009 · Browning 2012 · Calder 2017